Force-Guided Collaborative Control and Digital Twin of a Snake Robot for Cranial Bone Surgery
Adnan Munawar, Thomas Looi,, Eric Diller,, and Dale Podolsky
AI summary
Problem
Directly controlling high-degree-of-freedom snake robots in constrained cranial passages is difficult for surgeons, while traditional teleoperation reduces intuitive feedback and increases procedural risk.
Approach
The system translates surgeon-applied handle forces into precise motion along a planned trajectory, using a synchronized digital twin to provide real-time visualization and virtual fixtures that guide and constrain the robot's shape.
Key results
- Average positional errors of 6.79 mm (handle) and 7.13 mm (tool-tip) under guidance
- 5 N tangential force threshold optimally balances responsiveness and stability
- Successful follow-the-leader cutting on a 2 mm thick physical skull model
- Synchronized digital twin framework for pre-operative planning and intraoperative feedback
Why it matters
Enables safer, more intuitive human-robot collaboration for neurosurgeons performing minimally invasive cranial procedures, reducing reliance on isolated teleoperation.
Abstract
Tendon-driven snake-like robots have demon- strated potential for minimally invasive surgery, particularly in anatomically constrained regions. A previously developed system for cranial bone cutting integrated such a snake robot with an industrial manipulator, yielding a combined 9 degrees-of-freedom (DOF). However, direct operation of a high-DOF system is difficult for a human operator to control. Conventional robotic platforms often rely on teleoperation, which, while effective, physically separates the operator from the surgical site, reducing intuitive control and potentially increasing procedural risk. In this study, we introduce the first system that integrates a force-guided controller with a real-time digital simulator and full-shape virtual fixture (VF) to enhance human–robot interaction in cranial surgery. The controller enables shared autonomy, allowing the surgeon to apply input forces while the robot guides motion along a desired follow-the- leader (FTL) trajectory. The simulator provides synchronized visualization of the physical and virtual environment, offering both surgical planning insight and intraoperative feedback. Experimental results demonstrated that the proposed system facilitates collaborative control. The system shows improved accuracy and stability compared to manual operation, with average errors of 6.79 ± 3.27 mm at the handle and 7.13 ± 1.39 mm at the tool-tip, versus 13.61 ± 3.05 mm without guidance. The results also demonstrate that a 5 N force threshold offers the best balance of responsiveness and stability. Using this system, the robot successfully performed FTL cutting motion on a physical skull model with 2 mm bone thickness.